Mobile Phone Gesture Recognition Research Based On Inertial Sensor

The emergence of portable mobile devices has greatly changed human life. As oneof the very important means, the cell phone appears in front of people much morefrequently, meanwhile human beings interact with the phone everywhere. The contentof this paper is to get a new, interesting and more convenient human-machineinteraction: inertial sensor-based mobile phone gesture recognition.Input with keyboards and touch screen are two traditional methods for cell phones.To be different with these two mature ways, the inertial sensor-based mobile phonegesture recognition is an emerging research field, and this research covers signalprocessing, pattern identification and some other relevant aspects of the discipline. Ingenerally, in this method, the cell phone gets the users intention by reading the data ofthe MEMS inertial sensors (micro-accelerometers, micro-gyroscope, etc.) thatintegrated in the phone. Although some equipment has been used this method, theinertial sensor-based mobile phone gesture recognition is still in a relatively early stage,at the same, the complexity of the using environment, the difference of each gestureand the hardware of the device all means this method couldn’t be satisfactory inrecently.In this paper, based on the results of experience in related fields, it put forward afast and accurate gesture recognition method by using the built-in MEMS inertialsensors. Different from other gesture recognition systems, all the processes of inertialsensor-based mobile phone gesture recognition must be done by the cell phone itself,so, considering the phone can be bought, the core of this study is to research how touse the least system resources or calculate as little as possible to deal with the sensordata, as well as extract the gesture’s characteristics, and can provide the identificationresults immediately. Combining the existing resources of the mobile phone, gesturerecognizing methods by using accelerometer and comprehensive utilization ofthree-axis micro-accelerometer and three-axis micro-gyroscope are all studied. Andthen implant the recognizing algorithm into a cell phone, adjusting the algorithm, andfinally get a feasible inertial sensor-based mobile phone gesture recognition method.